okay hi everyone thank you for joining us for today's live on servicenow session on process optimization so just a quick Safe Harbor notice in case we do mention anything uh that's forward-looking uh please note that there might be some changes so it's only basically purchasing decisions on anything which we talk about in the future in case you're not aware this is part of a series of interactive events uh called live on servicenow so there Mike will share a link on the chat where you can access further events for the future quick bit of housekeeping so we've got time at the end for Q a but also please please put your questions into the Q a chat as well and we'll see if we can get to those along the way or we can answer them at the end so this presentation is going to be recorded and it will be shared on the servicedale community after the event you may get a short survey so we appreciate any feedback that you would like to give us so that we can improve uh future events so quick intro to myself my name is Alex Chaw I'm a product success manager here at servicenow my role is to help customers successfully adopt our idsm and devops products within the ABJ region so I've been with service now for almost nine years and I'm one of 200 or so Master architects in the ecosystem so I see we've got five participants with us today I just want to get a bit of an idea on everyone's experience on process mining so you know whether you know what it is at all uh whether you've seen it before whether you've used it before or maybe you you are an expert so just got a poll there if everyone can just uh I'll give you a couple of minutes just to respond okay um so it looks like four people responded out of the fix that we have uh and most of you don't know what process mining is uh one of you has seen a demo okay so hopefully we can explain to you a bit more about process mining today and I will also be going through a bit of a demo as well so in case you don't know our product is currently named process optimization and what it does is its function is to do process mining uh now there is some changes coming so in Vancouver I believe we will be renaming the product itself to process mining um to get rid of some of the confusion that they might be in about what process optimization actually is so I believe there will be changing the name soon but that doesn't really impact how the product works so we'll definitely the product is still going to be very much the same so what you see today so what is process mining so with everything we do there is a designs and a desired path in our mind on how something should work right so for example if you're in the park you'll see that there's there's designated areas for you know play there's paths that are made for people to walk in um and you know it's been designed by by a park Planner on how they think the park should best be used and it may be designed for things like efficiency maybe designed for things like completeness uh and to give everyone the best possible experience but in reality it doesn't always happen that way all right so you'll see here in this picture here you know lots of people are taking shortcuts uh they're not using the design path and this is really the same thing in in our business processes right so you'll you'll often see that your business processes uh might have an optimal path but in reality not everything follows that optimal path so so the problem with that is analyzing and improving business processes today is a very manual process and it can take on average about 12 weeks and costs upwards of you know 120 000 and there's still a high risk of error so in today um some of the methods that you might employ is analyzing reports which is a lot of manual work holding workshops which is time consuming because you have to get everybody in the room um for for hours or days you might have to hire external Consultants so-called experts which is also costly and then after all that's done you have to say internal alignment on everything and with all the time it takes to do all of that your target has probably moved by the time you actually get agreement to do some optimization so this is where service now is uh in platform process mining solution comes into it so currently called process optimization it gives you it gives process analyst business analysts process owners team managers the ability to very quickly see how their processes are working within the tool so we have traditional analytics things like performance analytics and Reporting these are the types of questions that get answered and it helps you answer things like you know how many tickets did we work on which type of tickets how long did it take to close those tickets how did we perform against our Target did we improve our cycle time right so those are the traditional uh those are the answers that you get with traditional analytics but what's under the iceberg there there's a lot of other information that can be gained and you can gain a lot of information through process mining so refers as mining you get more insight into where the process is actually stuck where do we have unnecessary degree work there are things going around that Loop where are we wasting time you know what's the highest process variance and so on so quickly how does this all work so every record on the platform creates an order Trail so we're typically talking about tasks right so things like incidents problems changes uh any sort of task they all leave and all the trail when they move through the system when changes are made to the task so the first thing that you need is you need to have auditing enabled on the table that you want to mine so by default most of those task tables have auditing turned on so I think there are about 120 tables which already have all the thing turned on so you probably have what you really need to get started what happens then is we have a mining engine uh which prepares an event pipeline for the relevant audit data handles for the for all the number crunching basically so we combine that with a growing number of findings which guide customers through the relevant Insight through the process optimization workbench so these are these uh what I'm talking about there is is content hacks So within the content packs we have sort of common patterns and things that we see within the processes so there's various content packs for itsm there's some HR CSM and so forth so we're in those content packs we have those uh common findings that we see within that process area so those in along with the mining engine uh we uh we Leverage The predictive intelligence infrastructure um so if you don't know what that is is it's separate to your instance so it's offloading the processing uh of the the mining to these predictive inhalers and servers so that it doesn't affect the performance of your system so that's what basically happens packages things get packaged up against the senses of predictive intelligence server to do the mining and the results come back to your instance so it doesn't obviously send everything about the task it only sends the things which you define in your project configuration and we'll see a little bit more of that when we go into the demo so the other thing is there are pre-built Integrations with other parts of the platform like performance analytics and benchmarks so basically if you're looking at a performance analytics graph you can hit a button and it will mine the uh the associated information with that performance indicator um I might show you a bit of that later as well and similarly with benchmarks right so if you're looking at a benchmark you can jump on to process optimization and it'll show you the process visualization associated with that benchmark from there once you have done your analysis you can also directly create a continual Improvement initiative from within the the process optimization workspace all right keep going just quickly check if there's any q a uh in the chat or no questions yet Alex okay great so before we jump into the demo quick overview of the process right so optimizing a process within service now we break it down into four phases so detect analyze Improvement monitor so how do you detect an open anomalies it's probably with a lot of capability that you have already say so you may have SLA notifications or breaching slas you might have performance analytics targets and thresholds Spotlight for example so it's uh you're probably getting idea of a anomaly anomalies from your system today already so what we then add to that we process optimization if you're then able to understand the process in more detail and find records so this is where a process optimization comes into it right so you get your process Maps you'll be able to run bottleneck analysis recourse analysis variation analysis and it also automatically runs automation Discovery I'll talk about a little bit later [Music] and then from there as I said you can create continual Improvement initiatives and you can use the other capabilities within the platform to help improve on those process areas so for example you might be appoint you to Virtual agent or predictive intelligence automation engine RPA or maybe it's just coaching right maybe you find that there's a bottleneck with a team that needs coaching so from then on you can Baseline and continually improve so with that I'm gonna jump into a demo so let me try and get that up so Mike can you see my servicenow instance yes so I'm going to start here so basically I'm logged in as somebody who has uh uh process optimization role to come into this process optimization workspace which is here so the first screen you're presented with is process projects um so yeah so oops so the the first screen represented here with is process projects um so this is where you sort of Define a scope around what you're analyzing so this is one that we've previously defined this one is for incidents so what I might do is I might edit the project so you can do it from here or you can do it from here if you want to create a new project and this is really you know we're defining the parameters of the process or processes that we're trying to analyze right so in this case I'm looking at incidents and problems I haven't added a goal but I couldn't tie it to a goal um oops that's not too great uh and then I'm also as you can also associate it with a with a kpi it's a performance analytics dashboard right so you know what are we trying to improve this is down here is where we're defining which processes we're looking at so we're looking at the incident table right there's a filter there uh active is false so we we've process mining it's it's generally best practice to my only mine close records right because open records could still be changing it's going to open this for a second so we're in that instant definition right you can Define your your conditions um to limit the things that you're gonna mine you also define activities and breakdown definitions so what are the key things that you want to see in your process map I want to see assignment group State pause bio problem ID I want to break it down by Business Service summer Group contact time uh so on and so on so the neck and then I'll also got the child table there as well right so this is linked to problems so uh we're also going to see the connections to problems within the process map so from here once all that's done um actually I should also mention this so process findings uh like I said with those content packs we give you uh common findings that we have found Within These processes so things like multi-hop issues uh bad intake quality long closed time so these patterns or findings have been predefined in our concert packs so when you you can you can associate that with your project [Music] I mean you can modify these as well so these are defined and you can actually go in and change you know what you might consider as a solution rejection for example um so with that once you've defined that you can then generate the model you can generate a quick sample model just to check that everything works um or you can generate a full model which will then give you your process map going back into the workspace so this is one where I've pre-generated before I'm going for that to load so you'll see the you know the parameters uh you can see the parameters along here or what it is that we mine so we mind the incident table 47 000 records in total six six point six thousand routes amongst those 47 000 records average duration time seven days standard deviation and so on and so forth what you'll also see along the top here is so summary in the insights is the first tab so this will give you this will will surface up things like Improvement opportunities uh it will show the goals and kpis you've tied the two I don't know why it's taking this long to load [Music] just quickly trying to do that again so this is a demo instance so it is quite slow okay that's better yeah so this is uh you know when we were defining a project before we tied it to a kpi so this is the kpi that we tied it to the the incident uh average time to resolve instance what you'll also see down here is the Improvement some improvement opportunity so these are the things that match those findings that we added to the project all right and we'll talk a bit more about that in a second it also gives you a variation analysis so basically you know what were the routes that the tickets that we analyzed went through so this is demo data so it might look a bit weird right created a problem attached uh it usually starts with new resolved closed but it's just that uh this is demo data so some of this stuff was inserted but you can start to see you know the different ways that records are moving through in this case the incident process right so I've got this ordered by highlights here I could be sorting this by the most records so let's set roads so most of the records 2.2 000 of the 47 000 are going across this path and you know I can click this and it will open this up in the next tab in the analyst workbench and it'll be filtered for these records but um we'll do that in a second [Music] so the other thing it presents you right on the summary in the insights page is the bottlenecks so bottlenecks are really you know State transitions where potentially there might be you know my beginning stuck so at the moment this is looking at overall Transitions and unique occurrences um let's say for example I could start going assignment group and total duration right so this is telling me you know uh there's there's 50 unique occurrences of support going to ID support Americas and on average it took four weeks before that transition to happen so you know you can start looking at and going oh hang on is it sitting with support I.T for too long here and then you can start drilling down into those specific records and doing investigation into that so like I said I could click this and it'll jump into the analyst workbench uh with that filter so let's do that now we'll come back to Improvement opportunities so yeah so this is you're going to get straight in value straight out of the bat you're going to get inside straight out of that so you know these findings are really telling me hey I've already found you know once here this is the first one long idle time how many how long before an agent started working after an assignment there's a potential impact there of so many years the average duration of those you know there's 21 000 records where the average duration was five days and then from here you can start leveraging your some of the itsm program buildings so the predictive intelligence stuff so we can quickly see oops I didn't mean to do that so from here you can quickly do a cluster analysis and automated recalls as well so um these will you have to run these when you want them done these are ones that are run already because it can take a little bit of time to do the cluster analysis so if I can click that it's loading but basically what's that what that is is that's that's gonna use predictive intelligence cluster those records and look for the biggest clusters and then you're selling you're starting to be able to quickly see trends I think that's twice okay there we go so you know the biggest the concept here it's found is issue that's the biggest cluster um the next cluster is application access user right so it's it's doing unstructured um it's doing analysis of unstructured data if you if you know what the the cluster framework is uh but if you don't that's that's what it is um it's doing it's it's probably analyzing the short description and it's coming up with okay there's 1 000 records where this is the theme or the concept right so quickly what I'm seeing is hey we've got long idle times um but a lot of those you know are related to application access user access uh PC is not working HDMI ports so this already starts to give me an indication that potentially I could be creating some self-service here right so then it doesn't have to get wet it doesn't have to wait to be assigned to someone and we could be saving potentially you know average of five days for each of those thousand records I did it again oh that's not quite there yet um yeah so that's some insight that you're getting straight away the other thing you can do is run some automated records so as I said the instance is a bit slow so it takes a bit of time to load and I've done the same thing again okay sorry oops okay of course come up yep so the records analysis you know again it's it's it's looking for patterns looking for Trends and it's telling me that you know 2000 or 2 000 of Records 2 700 records were assigned initially to the service desk all right so there may or may not be some information that's that's useful here right um let's stop doing that but um I think the other two are probably more useful in this case right so it's telling me that hey this specific person had 900 cases uh that were initially sent to him and it was 335 times more likely than than any other to have a long idle time so then you can start you know thinking about oh is there is there a problem with how this person is doing his work or their work do we need to you know put some coaching in place because for the coaching in place and see see if we can improve his efficiency oh so now what I'll do is I might jump into this one so instance waiting caller info so because I jumped in from there you will have uh it'll initially be filtered for those specific records yeah so one one filter has been applied instance a waiting caller info as I said if I jumped in from somewhere else the fields have would have been applied differently but this is one I jumped into so from here you start to see breakdowns and variation analysis so I can break this down by channels so this is you know initially when we defined the project we defined what we were going to be able to break it down by so this starts to come up here so I could for example start looking at those that are just self-service right which is probably you know in in a real context is probably where I would start because it's saying it's taking you know there's 1 700 records here where it's taking two weeks and four days on average for self-service that that sounds like that is uh you know something I would want to investigate so I could start doing that apply the filter there and then I'd just be seeing the process path just for those records right um now the other thing you'll notice is this is currently showing there are 682 routes this isn't showing all the routes all right so I can expand I can expand the connections here to show all the different routes all right and that's usually it's a bit too much to see everything uh you're probably looking at the 80 20 Rule and looking at hey 80 of the stuff is going to be oops in 20 of the routes in this case 30 . I'm just going to clear this filter for a second we'll just go through all of them all the channels that is okay and the other thing you see here is also again variation uh variation analysis right so if I bring this up a bit if I highlight over here you can see down the bottom uh you can see the route highlighted and you can see the transitions down the bottom and again you can you can use this to filter down if I just wanted to see what these four particular records were doing I could filter that so [Music] um that's about right still no questions nothing in the Q a Mike uh nothing yet Alex feel free to put any questions in the Q a or the chat if you have any so let's zoom in a bit so you'll see on this process map there are some you'll see you know the this the activities in the states that we defined um and the lines represent the transitions between those activities and states and between in within the line You'll see two little boxes what that is is the the metrics so that represents you know 4 600 records Traverse this path uh and I typically did that straight away in zero seconds that's because it's it's the creation uh the dummy creation what we might look at here for example is you know awaiting awaiting caller info so waiting caller info uh 1 800 records went to in progress and typically it's it was about two days between that trans transaction uh and you can change those metrics so we're looking at unique occurrences you could change that to Total occurrences um we can change that to duration I think let's bring that down again so you can change you can change these Global metrics and that's going to show you that that'll affect this these boxes here so that you can see uh quite quickly you know what what the things that look at let's go back to another generation okay I have one question in the Q a is the process optimization capable of analyzing which CIS in cmdv table has the most incidents slash events raised if so what the variation analysis available to use it in the application of CI perspective uh I'm gonna say it's not it's not really the it's not really the yeah it's not really it's not really the purpose or process optimization process optimization you know to to tell you which CIS have the most events it's more about your process as a whole and how things are traversing through your process um so you know if you wanted to know which which the eyes had the most events that would be a much that would be a really easy report just to run so I wouldn't use process optimization to find out information I I think that info is available in some of the out of the box uh reports and dashboards that I've seen so yeah you don't need po to get that info it's more about it's more about analyzing things like this right so we can see here uh this path here has gone from in progress to a waiting caller info so a lot of stuff well 3900 records from uh uh oh actually I'll look at look here this probably gives you better info so occurrences right total 4 600 records went along that path from in progress to waiting caller info 3895 of those were unique which means there were some which went more than one time right multiple times the max repetitions is nine so there was one record or there was some records where it went in progress to a wedding caller nine times so I can actually start delving down into the information on or you know investigating that so if I wanted to so this is currently incident distribution by duration I could do it by repetitions I go this nine so that nine repetitions it tells me that one record went through nine homes I can actually look at that one record so this is the path that that one record went through um you know this is the steps oops you can see what happened there in progress of wedding core in Progressive waiting process in progress awaiting caller um I can even if I wanted to go view the record so that's how powerful this is it didn't actually get down into you know a specific record within your whole process so I'm going to I won't do that let me go back to clear this chain oops [Music] let's go back to instance surfaces again this is all incidents um incidents to awaiting caller info let's go back to here where we were um yeah so I was able to you know dribble down into what was the record that had the nine requisitions um I can see most of the records out of that four thousand three and a half thousand you know talk about one day to get additional info from from the customer from the caller which is you know not so bad right so it can give you good indicators as well as bad indicators if I wanted to I could also show these records from here um do root cause analysis from here do some cluster analysis as well right so you know maybe this is an indicator that hey for email issues it can't access so we are if there are potentially some is it is potentially self-service is there any catalog item or something you know maybe we're not asking the right questions because if we're having to go back 790 times um to the customer to ask about this stuff right so then you could you start having that insight into being able to improve those processes that's catalog items so if you do find something like that you know one thing you can do is actually create a note so this is one I created earlier um you can create a new note you know tell somebody else to so somebody else to have a look at this and then you can okay you can also attach the snapshot all right and so when they come in they'll come to this um specific map and be able to see what happens and what you're asking them to look at as I said if you do find an initiative right you do find an improvement you can then go and create a continuous Improvement initiative I created one before so you can do it straight from here or you can link it to an existing initiative if that one already exists if you have a automation Center you can also start requesting automation all right so that's that's something separate that you can do we'll put in a request and then you know someone can come along with the platform capabilities and potentially automate some of that away as well I'm now going to jump over to unless there or any other questions I'm gonna jump over to automation opportunities so this is another tab that automatically gets it only gets run once you do a process mine or once you you know do the mining so this is the capability that you don't need to have process optimization for so you just need icsm Pro and it'll use predictive intelligence users with clustering capability to tell you so basically for those that don't know what this is it it's again based on the table that we've defined it's done some analysis predictive intelligence machine learning and it's matched up against a known taxonomy and iitsm taxonomy and selling you you know what are the potential uh opportunities for you to potentially deflect or automate away so you know could I build a virtual agent conversation to to help troubleshoot Hardware right and that will potentially you know help me get rid of the 3700 tickets so this is something that's automatically run with process optimization but if you don't have process optimization you can actually still still run this as well and from and if you do run this so if you you came in came in from this angle you can also open the process map for this as well so the products are tightly integrated uh one thing I forgot to mention is so once you've done that mining you can actually save different versions so you know maybe you maybe you have some improvements that you make to this process and then you've run it again in a month's time to see if there's been any change so what you can do is actually start to compare versions so this is the version that we have I'm going to compare to a version that I had earlier right and then you can start to see um the differences of course it's just refreshing all right so that's what I wanted to show in the demo today unless there's any other questions I will switch back to the panel point foreign at this time Alex so these are just some of the common findings and optimization opportunities that we have found with the product so you know some of these have been defined as patents that you will get when you run the process optimization and these are other some of them may be other things that you find as you go through your process now these are some of the common things that that we have found in our experience at least they'll be there in in the deck when when you get them so uh real live use case as uh sorry one second a real live use case is um servicenow so service now ran process optimization um on RIT support and you know we saved up to 141 days right so what we found in this particular transition here 141 days was that things were going to the wrong team so it was going from I think I'd uh I was saying it was going to the wrong team and basically that team wasn't able to see the record and therefore was just sitting there for 141 days so you know that's that that's where we were able to use process optimization to identify that and have the evidence to support that and and quickly get that fixed okay I've got a question can you share a link to this recording in previous recordings yeah so this recording will be posted into the community Forum um I believe we we should have a link for you if not we'll get that to you once um uh once the meeting is finished I think it takes a couple of days for the recording to become available and um and and that will get sent out to all attendees thanks Mike all right so process optimization so it isn't uh it isn't a free product you do it is a it is a licensed product but customers can actually try it for free so you can activate uh you'll see the plugins down the bottom there so process optimization and then the itcm process optimization content though so you can activate these plugins uh in a non-production instance without any licensing that will be restricted to 4000 records so you can try some of this yourself and see you know see the power for yourself so you can do that today you can go in and activate those and and get that running so from a production standpoint a separate license is required before it can it be activated on production and then once you have that activated it you can mine the Full Table uh or the tables that are part of your entitlement so um you know if you have itsm you can only mind the itsm tables and CSM tables if you have the CSM licensing for example foreign so now that you've seen it um you understand a bit more about it I'd like to see which process you're interested in in mining if you had this I've got three out of five okay that's a couple more answers [Music] okay I might end up there um yes so incident and change so yes those are the two most um I would say the the two biggest areas that we see people using this for and that we have content for as well actually so we have content packs the content pack includes uh stuff for instance and change and those those are the biggest areas where we think that improvements can be made so good to good to see that lines up oh all right so if you want to learn more about process optimization there's a free process optimization Essentials course so you can take that at now learning dot your learning I think Mike will actually pay some of these in the link sorry in the chat as well so you've got the I've got the links handy for you we also have the community forum for for process optimization so it's got the the basic stuff for getting started and any information that you need can be asked on the Forum there's an FAQ that's on the Forum as well if you want to find out more information more detailed information about process mining some of the frequently asked questions in there and I know there is also a process optimization Academy so that overview and demo that's that's available uh in the academy as well and our product manager Dan Grady I believe he's responsible for for process of civilization then he posts some really great stuff uh in the process optimization Academy so Mike will post that in the link in the chat thank you Mike and we've got a few minutes left for any questions um actually allow everyone to talk as well so if you want to come off mute and ask your question feel free to do so thank you Machu thanks for staying up late himanshu yes thank you all the way from Toronto awesome all right so if we have no other questions um thank you very much for your time so as you know we we have a number of offerings um that can help you if you're looking at the process optimization or or any other capability that you're looking to roll out and need support for you know we've got training and certification we've got expert Services if you you need someone to help you Hands-On and then we'll also got our impact offerings um where we've got you know success managers that can help you get the best out of your products thank you very much for your time uh no other questions we'll we'll end the webinar we'll stop the recording and hopefully we'll be able to send out the link to the materials and the recording uh soon after foreign